I am a member of the team Probabilités et statistique of Montpellier Institute Alexander Grothendieck (IMAG), and associate member of the team Contrôle de Qualité et Fiabilité Dynamique CQFD of Inria Bordeaux Sud-Ouest.
I defended the habilitation à diriger des recherches in July 2013. My thesis is titled:
Contributions à l'estimation et au contrôle de processus stochastiques
PhD supervision ↑
Tiffany Cherchi (2017-2020) ↑ shared supervision 50% with François Dufour, funding requested from Cifre Thales Optronique
Automated optimal fleet management policy for airborne equipment
This project is concerned with the analysis and the study of the Markov Decision Processes (MDPs). It focuses on the modeling of a fleet of equipment with several levels of performance, which can be broken down or required for mission-oriented tasks. Thales wishes to put in place an optimized maintenance policy for the equipment to guarantee the availability of the equipment and the proper deployment of missions while minimizing maintenance costs. This problem can be seen as a fleet management problem. One of the main objectives will be to develop theoretical and numerical tools for the solution of such optimization problems.
Maud Joubaud (2016-2019) ↑ shared supervision 50% with Bertrand Cloez, funding UM and FEDER
Branching piecewise deterministic Markov processes, applications to cell biology
The aim of the thesis is to study new models of branching PDMPs to model the dynamics of cell populations. We will pay particular attention to the simulation and optimization of these processes. Two examples will be privileged: the division of Escherichia coli related memory problems and asymmetry of the division, and chemostat models with possible applications to depollution.
This thesis was defended on 25 June 2019. Maud is presently working as a high school teacher.
Alizée Geeraert (2014-2017) ↑ shared supervision 50% with François Dufour, funding Cifre Thales Optronique
Optimal control of piecewise deterministic markov processes with application to maintenance
The proposed research topic aims to study and analyze stochastic optimal control problems for PDMPs. This work we will focus on impulse control problems when the controller punctually intervenes on the process by moving it to a new point of the state space at a time specified by the controller. Mathematically, this is to build an optimal control strategy defined by a sequence of time and stop points of the state space, inducing new jumps to the process in order to optimize function cost defined from a functional process. The main purpose of this thesis is to analyze this type of optimization problems in particular characterize the existence and shape policies? Epsilon-optimal and develop digital tools for the simulation of these strategies.
This thesis was defended on 6 June 2017. Alizée is now teaching at the military school of Brittany.
Christophe Nivot (2013-2016) ↑ shared supervision 50% with François Dufour, funding Airbus Defence & Space/Région Aquitaine
Modélisation et optimisation d'une chaîne de montage par les processus markoviens décisionnels
This work deals with Markov Decision Processes (MDP). MDPs are a general family of controlled stochastic processes, which are suitable for the modeling of several sequential decision-making problems under uncertainty. They arise in many applications, such as engineering, computer science, telecommunications, finance, etc. The main aim of this work is to develop theoretical and numerical tools to solve such problems, by using dynamic or linear programming. Results are applied to the optimization of the assembly line of the new European launcher in collaboration with Airbus Defence & Space.
This thesis was defebded on 19 May 2016. Christophe is presently working as a high school teacher.
Adrien Brandejsky (2009-2012) ↑ shared supervision 50% with François Dufour, funding Astrium
Méthodes numériques pour les processus markoviens déterministes par morceaux (PDMP)
PDMP's have been introduced by M.H.A. Davis as a general class of non-diffusive stochastic models. PDMP's are hybrid Markov processes involving deterministic motion punctuated by random jumps. In this thesis, we develop numerical methods that are designed to fit PDMP's structure and that are based on the quantization of an underlying Markov chain. We deal with three issues : the approximation of expectations of functional of a PDMP, the approximation of the moments and of the distribution of an exit time and the partially observed optimal stopping problem. In the latter one, we also tackle the filtering of a PDMP and we establish the dynamic programming equation of the optimal stopping problem. We prove the convergence of all our methods (most of the time, we also obtain a bound for the speed of convergence) and illustrate them with numerical examples.
This thesis was defended on 2 July 2012. Adrien is presently Finacial Risk Analyst at Nomura London.
Karen Gonzalez (2007-2010) ↑ shared supervision 50% with François Dufour, funding MENRT
Contribution à l'étude de processus markoviens déterministes par morceaux. Etude d'un cas-test de la sûreté de fonctionnement et Problème d'arrêt optimal à horizon aléatoire
In a first part, PDMPs are used to compute probabilities of top events for a case-study of dynamic reliability (the heated tank system) with two di erent methods : the rst one is based on the resolution of the differential system giving the physical evolution of the tank and the second uses the computation of the functional of a PDMP by a system of integro-differential equations. In the second part, we propose a numerical method to approximate the value function for the optimal stopping problem of a PDMP. Our approach is based on quantization of the postjump location and inter-arrival time of the Markov chain naturally embedded in the PDMP, and path-adapted time discretization grids. It allows us to derive bounds for the convergence rate of the algorithm and to provide a computable epsilon-optimal stopping time.
This thesis was defended on 3 December 2010. Karen is now Defendability and Safety engineer at EADS-APSYS.
Publications ↑
Book ↑
- [L1] B. de Saporta, F. Dufour, H. Zhang, Numerical methods for simulation and
optimization of piecewise deterministic Markov processes : application to reliability, Mathematics and statistics series Wiley-ISTE, 2015 page web éditeur
Patent ↑
- [B1] D. Bihannic, C. Baysse, B. de Saporta, F. Dufour, A. Gégout-Petit, J. Saracco, Procédé de maintenance d'un é quipement, Demande de brevet : France N1303026 N/Ref. : 068689 FR MPH/MAG, Thales, Inria, 2013
Publications in international peer-reviewed journals ↑
- [A26] B. Cloez, B. de Saporta, M. Joubaud Optimal stopping for measure-valued piecewise deterministic Makov processes, Journal of applied Probability 2020, to appear arXiv:1809.04824
- [A25] B. Delyon, B. de Saporta, N. Krell, L. RobertInvestigation of asymmetry in
E. coli growth rate, CSBIGS 2018, pp 1-13 arXiv:1509.05226
- [A24] A. Cleynen, B. de Saporta Change point detection for piecewise deterministic markov processes, Automatica, 2018, pp234-247 arXiv:1709.09467
- [A23] H. Zhang, B. de Saporta, F. dufour, D; Laneuville, A. Nègre Stochastic Control of Observer Trajectories in Bearings-only Tracking with Acoustic Signal Propagation Optimization, IET Radar, Sonar and Navigation, 2018, pp 112-120 arXiv:1703.09924
- [A22] EF. Costa, B. de Saporta, Linear minimum mean square filters for markov jump
linear systems, IEEE Trans on automatic control, 2017, pp3567-3572 arXiv:1601.00772
- [A21] B. de Saporta, F. Dufour, C. Nivot Partially observed optimal stopping problem for discrete-time Markov processes, A quarterly journal of Operations Research 2017, to appear arXiv:1602.04609
- [A20] B. de Saporta, F. Dufour, A. Geeraert Optimal strategies for impulse control of piecewise deterministic markov processes, Automatica 77, 2017, pp219-229 arXiv:1603.07859
- [A19] B. de Saporta, EF. Costa, Approximate Kalman–Bucy filter for continuous-time
semi-Markov jump linear systems, IEEE Trans on automatic control 61(8), 2016, pp 2035-2048 arXiv:1409.2631
- [A18] B. de Saporta, A. Gégout-Petit, L. Marsalle, Statistical study of asymmetry in cell lineage data, Computational Statistics & Data Analysis 69, 2014, pp15-39 arXiv:1205.4840
- [A17] B. de Saporta, A. Gégout-Petit, L. Marsalle, Random coefficients bifurcating autoregressive processes, ESAIM P&S eFirst 5, 2013 arXiv:1205.3658
- [A16] A. Brandejsky, B. de Saporta, F. Dufour, Optimal stopping for partially observed piecewise-deterministic Markov processes, Stochastic Processes and their Applications 123(8), 2013, pp3201-3238 arXiv:1207.2886
- [A15] B. de Saporta, H. Zhang, Predictive maintenance for the heated hold-up tank, Reliability Engineering and System Safety, 115, 2013, pp82-90 arXiv:1207.4866
- [A14] A. Brandejsky, B. de Saporta, F. Dufour, Numerical method for expectations of piecewise-determistic Markov processes, CAMCoS 7(1), 2012, pp63-104 arXiv:1105.0839
- [A13] B. de Saporta, A. Gégout-Petit, L. Marsalle, Asymmetry tests for Bifurcating Auto-Regressive Processes with missing data, Statistics and Probability Letters 82(7), 2012, pp1439-1444 arXiv:1112.3745
- [A12] B. de Saporta, F. Dufour, Numerical method for impulse control of Piecewise Deterministic Markov Processes, Automatica 48, 2012, pp779-793 arXiv:1011.5812
- [A11] B. de Saporta, F. Dufour, H. Zhang, C. Elegbede, Optimal stopping for the predictive maintenance of a structure subject to corrosion, Journal of Risk and Reliability, 226(2), 2012, pp169-181 arXiv:1101.1740
- [A10] A. Brandejsky, B. de Saporta, F. Dufour, Numerical methods for the exit time of a piecewise-deterministic Markov process, Advances in Applied Probability 44(1), 2012, pp196-225 arXiv:1012.2659
- [A9] B. de Saporta, A. Gégout-Petit, L. Marsalle, Parameters estimation for asymmetric bifurcating autoregressive processes with missing data, Electronic Journal of Statistics 5, 2011, pp1313-1353 arXiv:1012.2012
- [A8] B. de Saporta, F. Dufour, K. Gonzalez Numerical method for optimal stopping of piecewise deterministic Markov processes, Annals of Applied Probability 20(5), 2010, pp1607-1637 arXiv:0903.2114
- [A7] B. Bercu; B. de Saporta, A. Gégout-Petit Asymptotic analysis for bifurcating autoregressive processes via a martingale approach, Electronic Journal of Probability 14(87), 2009, pp 2492-2526 arXiv:0807.0528
- [A6] B. de Saporta, Tail of the stationnary solution of the stochastic equation Y(n+1)=a(n)Y(n)+b(n) with Markovian coefficients, Stochastic Processes and their Application 115(12), 2005, pp 1954-1978 pdf
- [A5] B. de Saporta, J.-F. Yao, Tail of a linear diffusion with Markov switching, Annals of Applied Probability 15(1B), 2005, pp 992-1018 pdf
- [A4] B. de Saporta, Tail of the stationnary solution of the stochastic equation Y(n+1)=a(n)Y(n)+b(n) with Markovian coefficients, C.R. Acad. Sci. Paris, Ser. I 340(1), 2005, pp 55-58 pdf
- [A3] B. de Saporta, J.-F. Yao, Tail of a linear diffusion with Markov switching, C.R. Acad. Sci. Paris, Ser. I 339(9), 2004, pp 643-646 pdf
- [A2] B. de Saporta, Y. Guivarc'h, E. Le Page, On the multidimensional stochastic equation Y(n+1)=a(n)Y(n)+b(n), C.R. Acad. Sci. Paris, Ser. I 339(7), 2004, pp 499-502 pdf
- [A1] B. de Saporta, Renewal Theorem for a system of renewal equations, Annales de l'Institut Henri Poincaré 39(5), 2003, pp 823-838 pdf
Book chapters ↑
- [C6] A. Cleynen, B. de Saporta, Change point detection for piecewise deterministic Markov processes, in Mathematical modelling of random and deterministic phenomena, Wiley, 2020, pp 73-96
- [C5] B. de Saporta, F. Dufour, Numerical method for control of piecewise deterministic Markov processes, in Statistical inference for piecewise deterministic Markov processes, Wiley, 2018, pp 147-172
- [C4] J.F. Aubry, G Babykina, N. Brinzei, S. Medjaher, A. Barros, Ch. Bérenguer, A. Grall, Y. Langeron, D.N. Nguyen, G. Deleuze, B. de Saporta, F. Dufour, H. Zhang, Projet ApproDyn : approches de la fiabilité dynamique pour modéliser des systèmes critiques., in Supervision and Safety of Complex Systems, N. Matta, Y. Vandenboomgaerde, and J. Arlat, editors, Hermes Sciences - Lavoisier, 2012, pp181-222
- [C3] J.F. Aubry, G Babykina, N. Brinzei, S. Medjaher, A. Barros, Ch. Bérenguer, A. Grall, Y. Langeron, D.N. Nguyen, G. Deleuze, B. de Saporta, F. Dufour, H. Zhang, The ApproDyn project: dynamic reliability approaches to modeling critical systems., in Supervision, surveillance et sûreté de fonctionnement des grands systèmes, N. Matta, Y. Vandenboomgaerde, and J. Arlat, editors, Wiley-ISTE, 2012, pp141-179
- [C2] B. de Saporta, F. Dufour, Numerical method for optimal stopping of piecewise deterministic Markov processes, in Modern trends in controlled stochastic processes, Luniver press, 2010, pp44-64
- [C1] C. Blanchet-Scalliet, R. Gibson Brandon, B. de Saporta, D. Talay, E. Tanré, Viscosity Solutions to Optimal Portfolio Allocation Problems in Models with Random Time Changes and Transaction Costs, in Advanced Financial Modelling, de Gruyter, 2009, pp 53-90 NCCR working paper
PhD and HdR theses ↑
- [Th2] B. de Saporta, Contributions à l'estimation et au contrôle de processus stochastiques., Université Bordeaux 1, 2013 pdf
- [Th1] B. de Saporta, Etude de la solution stationnaire de l'équation Y(n+1)=a(n)Y(n)+b(n) à coefficients aléatoires., Université Rennes 1, 2004 pdf
International conferences with proceedings ↑
- [PI15] B. de Saporta, F. Dufour, H. Zhang Dynamic optimization of maintenance policies for multistate systems , ESREL19 2019, Hannover, Germany
- [PI14] T. CHerchi, C. Baysse, B. de Saporta, F. Dufour, Optimal predictive maintenance policy for multi-componant system , ESREL19 2019, Hannover, Germany
- [PI13] EF. Costa, B. de Saporta, Precomputable kalman-based filter for Markov jump linear systems, 3rd International Conference on Control and Fault-Tolerant Systems 2016, Barcelona, Spain
- [PI12] C. Nivot, B. de Saporta, F. Dufour, J. Béhar, D. Bérard-Bergery, C. Elegbede, Modeling and optimization of a launcher integration process, ESREL15 2015, Zurich, Switzerland
- [PI11] EF. Costa, B. de Saporta, A numerical method for state estimation of continuous
time markov jump linear systems, 53rd IEEE Conference on Decision and Control 2014, Los Angeles, USA
- [PI10] H. Zhang, B. de Saporta, F. Dufour, D. Laneuville, A. Nègre, Optimal
trajectories for underwater vehicles by quantization and stochastic control, 17th
international conference on information fusion 2014, Salamanca, Spain
- [PI9] H. Zhang, B. de Saporta, F. Dufour et G. Deleuze, Dynamic reliability by using simulink and stateflow, Prognostics and System Health Management Conference 2013, Milano, Italy
- [PI8] C. Baysse, D. Bihannic, A. Gégout-Petit, M. Prenat, B. de Saporta, J. Saracco, Maintenance optimization of optronic equipment, Prognostics and System Health Management Conference 2013, Milano, Italy
- [PI7] H. Zhang, B. de Saporta, F. Dufour et G. Deleuze, Dynamic reliability: towards efficient simulation of the availability of a feedwater control system, NPIC-HMIT 2012, San Diego, USA
- [PI6] B. de Saporta, F. Dufour, H. Zhang, Predictive maintenance for the heated hold-up tank, PSAM11-ESREL12, 2012, Helsinki, Finland
- [PI5] A. Nègre, O. Marceau, D. Laneuville, H. Zhang, B. de Saporta, F. Dufour, Stochastic Control for Underwater Optimal Trajectories, IEEE-AIAA Aerospace conference 2012, Big Sky (MT), USA
- [PI4] A. Gégout-Petit, B. de Saporta, L. Marsalle, Analyse multi-arbres de l'asymétrie dans la division cellulaire, Journées de Statistiques JdS 2012, Bruxelles, Belgium
- [PI3] A. Brandejsky, B. de Saporta, F. Dufour, C. Elegbede, Numerical method for the distribution of a service time, ESREL 2011, Troyes, France
- [PI2] B. de Saporta, F. Dufour, H. Zhang, Approximation of the value function of an impulse control problem of Piecewise determinsitic Markov process, IFAC 18th world congress 2011, Milano, Italy
- [PI1] B. de Saporta, F. Dufour, K. Gonzalez, Numerical method for optimal stopping of hybrid processes, 3rd IFAC Conference on Analysis and Design of Hybrid Systems 2009, Zaragoza, Spain
International conferences without proceedings ↑
- [CI10] B. de Saporta, F. Dufour, H. Zhang, Dynamic optimization of maintenance policies for multi-component equipment , SIAM Conf on Control and its applications, 2019, Chengdu, China
- [CI9] B. de Saporta, F. Dufour, H. Zhang, Numerical method for impulse control of piecewise deterministic Markov processes, XVIème Colloque Franco-Roumain de Mathématiques Appliquées 2012, Bordeaux, France
- [CI8] A. Cleynen, B. de Saporta, Optimal stopping for change-point detection of piecewise deterministic Markov processes , SIAM Conf on Control and its applications, 2017, Pittsburgh, USA
- [CI7] C. Baysse, B. de Saporta, A. Gégout-Petit, Numerical method for optimal
stopping of PDMP and maintenance optimization, International Workshop on Applied
Probability 2014, Antalya, Turkey
- [CI6] B. de Saporta, A. Brandejsky, F. Dufour, Optimal stopping for partially observed piecewise deterministic Markov processes, XIème Colloque Franco-Roumain de Mathématiques Appliquées 2012, Bucarest, Romania
- [CI5] B. de Saporta, A. Gégout-Petit, L. Marsalle, Limit theorems for bifurcating autoregressive processes with missing data, 16th Informs Applied Society Conference 2011, Stockholm, Sweden
- [CI4] B. de Saporta, F. Dufour, K. Gonzalez, Numerical method for optimal stopping of piecewise deterministic Markov processes, Symposium on Optimal Stopping with Applications 2009, Turku, Finlande
- [CI3] B. de Saporta, C. Blanchet-Scalliet, R. Gibson, D. Talay, E. Tanré, Optimal portfolio allocation under transaction costs, 31st conference on Stochastic Processes and Their Applications 2006, Paris, France
- [CI2] B. de Saporta, C. Blanchet-Scalliet, R. Gibson, S. Rubenthaler, E. Tanré, D. Talay, Technical analysis compared to mathematical models under misspecification, AMAMEF conference Numerical Methods in Finance 2006, INRIA Rocquencourt
- [CI1] B. de Saporta, Renewal theory for a system of renewal equations, Second International Workshop on Applied Probability IWAP 2004, Université du Pirée, Greece
WoInternational workshops ↑
- [W4] B. de Saporta, Asymétrie et mémoire dans la division cellulaire, 5e Rencontres Scientifiques Sherbrooke-Montpellier 2015, Sherbrooke, Canada
- [W3] B. de Saporta, A. Brandejsky, F. Dufour, Numerical method to compute the law of exit times for piecewise deterministic Markov processes, Hitting times and exit problems for stochastic models 2013, Dijon, France
- [W2] B. de Saporta, F. Dufour, Numerical method for optimal stopping of piecewise deterministic Markov processes, Modern trends in controlled stochastic processes/span> 2010, Liverpool, UK
- [W1] B. de Saporta, B. Bercu, A. Gégout-Petit, Asymptotic behavior of bifurcating autoregressive processes, Mathematical models for cell division 2009, Paris, France
National conferences with proceedings ↑
- [PN5] C. Baysse, D. Bihannic, B. de Saporta, A. Gégout-Petit, M. Prenat, J. Saracco, Optimisation de la maintenance d'un équipement optronique, Journées de Statistiques JdS 2013, Toulouse
- [PN4] C. Baysse, D. Bihannic, B. de Saporta, A. Gégout-Petit, M. Prenat, J. Saracco, Modèle de Markov caché pour la détection d'un mode de fonctionnement dégradé d'un équipement optronique, Lambda-Mu 18, 2012, Tours
- [PN3] H. Zhang, B. de Saporta, F. Dufour, G. Deleuze, Fiabilité dynamique : simulation d'un système de régulation du niveau d'eau dans un générateur de vapeur, Lambda-Mu 18, 2012, Tours
- [PN2] B. de saporta, F. Dufour, H. Zhang, C. Elegbede, Arrêt optimal pour la maintenance prédictive, Lambda-Mu 17, 2010, La Rochelle
- [PN1] A. Gégout-Petit, B. de Saporta, L. Marsalle, Analyse asymptotique des processus autorégressifs de bifurcation avec données manquantes, Journées de Statistiques JdS 2010, Marseille
National conferences without proceedings ↑
- [CN2] B. de Saporta, Queue de la solution stationnaire d'une équation récursive aléatoire à coefficients markoviens, Colloque Jeunes probabilistes et statisticiens 2004, Aussois
- [CN1] B. de Saporta, Queue de la solution stationnaire d'une équation récursive aléatoire à coefficients markoviens, Journées MAS 2002, Grenoble
Technical reports ↑
- [R7] H. Zhang, B. de Saporta, F. Dufour, Filtrage et contrôle optimal stochastique, application à l'optimisation de trajectoire, Rapport technique de CQFD pour DCNS 2013
- [R6] J.F. Aubry, G Babykina, N. Brinzei, S. Medjaher, A. Barros, Ch. Bérenguer, A. Grall, Y. Langeron, D.N. Nguyen, G. Deleuze, B. de Saporta, F. Dufour, H. Zhang, Projet APPRODYN : APPROches de la fiabilité DYNamique pour modéliser des systèmes critiques, Rapport final pour le GIS Supervision, surveillance et sûreté de fonctionnement des grands systèmes, 2012
- [R5] H. Zhang, B. de Saporta, F. Dufour, Contrôle optimal stochastique, application à l'optimisation de trajectoire en 3D, Rapport technique de CQFD pour DCNS 2012
- [R4] H. Zhang, B. de Saporta, F. Dufour, Contrôle optimal stochastique, application à l'optimisation de trajectoire, cas multicible, Rapport technique de CQFD pour DCNS 2011
- [R3] H. Zhang, B. de Saporta, F. Dufour, Contrôle optimal stochastique, application à l'optimisation de trajectoire, Rapport technique de CQFD pour DCNS 2010
- [R2] R. Azaïs, B. de Saporta, F. Dufour, A. Gégout-Petit, M. Touzet, Aspects aléatoires de la propagation de fissures, Rapport technique de CQFD pour EADS-Astrium 2009
- [R1] B. de Saporta, F. Dufour, A. Gégout-Petit, H. Zhang, Modèles stochastiques pour la propagation de fissures, Rapport technique de CQFD pour EADS-Astrium 2008